基于红外测温技术的变电设备故障预警分析

Fault Warning Analysis of Substation Equipment Based on Infrared Temperature Measurement Technology

本文简要介绍了红外测温的基本原理,分析了变压器中的红外测温方案,提出了一种MIV和BRBP神经网络的变压器红外故障诊断方法,利用红外测温方式,获取了不同环境温度及运行状态下变压器中元器件温度数据。将此参数作为故障诊断模型的初始输入变量,经过MIV算法简约参数输入至BRBP神经网络,进行故障评估和诊断。结果表明:相对于传统的BRBP神经网络,本文设计的基于MIV和BRBP神经网络模型诊断方法极大简化了数据训练的数据量并解决了数据收敛的困难,因此效率更高,用时更省。证实了此方法的可行性。

This paper briefly introduces the basic principle of infrared temperature measurement, the analysis of the infrared temperature measurement scheme of transformer puts forward the transformer infrared fault diagnosis method of MIV neural network and BRBP neural network, it obtained the transformer in different environmental temperature and the running status of components in the temperature data by using the infrared temperature measurement method, This parameter is used as the initial input variable of the fault diagnosis model, and it is input to the BRBP neural network through the simple parameters of the MIV algorithm to evaluate and diagnose the fault. The results show that compared with the traditional BRBP neural network, the MIV and BRBP neural network model diagnosis method designed in this paper greatly simplifies the data volume of data training and solves the difficulty of data convergence, so it is more efficient and saves time. The feasibility of this method is confirmed.